Radiical Computing - Leaders in a Liquid Cooled World!

Home
Ask a Question
Secure Orders
Latest News
Reviews
Projects
Customer Gallery
Guides
FAQ
Live Testing


Copper Plate
Fans
Filling Kits
Fittings
Fluid XP
Kits
Monitoring
Pumps
Radiators
Reservoir
Shrouds
TIM
Tubing
Water Blocks
Glossary
Legal Stuff
Testimonials
Links
About Us
Contact


Aspire HTPC Project Three

Project Date Log February 18, 2008

Performance;

We have used three broad categories to establish the performance of our water cooled Aspire HTPC system.

The three categories are;
Acoustic level....... the volume of sound generated by the system.
Thermal performance....... what sort of temperatures to expect for this system
Processing performance..... the old bang for buck data.

Acoustic Performance

Okay so how loud is the system. Does it interfere with normal TV or other multimedia applications. The answer is "well it did sort of". To clarify the point being made here its necessary to look backwards at the development of the system.

Our aim was to produce as quiet a system as possible. The problem is of course that as you silence one moving part in the system you uncover a previously muted sound emanating from another moving part.Sort of like peeling off the layers of an onion. We started out with the power supply which produced an audible whisper probably not worthy of attention in a normal system but here it stood out. Solution.... out comes the Aspire PSU and in goes a silent PSU. Then it was the hard drive ticking over when multitasking which became noticeable. There was not much that could be done about this really due to the case size constraints. We solved the impasse by moving the hard drive lower down inside the case and the sound level was reduced to the point that it was hardly discernable. Finally we took on the DDC pump and insulated it from the motherboard tray with a thermal pad. We achieved two solutions by doing this. First the sound level was dramatically reduced and secondly the pump cooling improved by virtue of using the whole of the aluminium case as a very large passive heatsink. The final step was to limit the single radiator fan to 5 volts which brought the sound level to near silence. We were happy at this point with our creation and moved on to the next task thermal performance.

Thermal Performance

Before we start into our CPU thermal performance data it is worthwhile to have a look at our HTPC CPU choice the Intel 45nm cored E8400. We were attracted to this CPU over the older 65nm parts because of the number of positive references on various forums espousing how cool running and easily overclocked this CPU can be.

Sadly at the time of writing this the current temperature recording software monitoring thermal output is proving itself to be very inaccurate when tested on this CPU. This is not something unique to our E8400 CPU. There are a variety of reports from various reviews and over clocking forums indicating the difficulty in getting accurate temperature recording for the E8400. At the time of writing our temperature readout data normally provided by Core temp is proving to be useless for this CPU giving readings that are not realistic. Speedfan 4.33 is in much the same position although the accuracy difference is somewhat less than Core Temp (see the temperature figures given in red below). This has left us with very few software based options so we have returned to the thermal measurement methodologies mentioned below.

With the various software based measurement shortcomings in mind we have returned to measuring the delta/T water to ambient air values so that an accurate comparative value can be obtained. After a short period of run in to settle the TIM joint the stock ambient to coolant (ATC) temps were obtained as follows;

At Idle and default value of 3Ghz. Single Antec Tri Cool fan set on lowest setting.

Ambient Air; 24.80 degrees
Water Temperature; 27.20 degrees
D/t; 2.40 degrees
Speedfan 4.33 Reading; 39.40 degrees

At Load and default value of 3Ghz. Single Antec Tri Cool fan set on lowest setting.

Ambient Air; 27.60
Water Temperature; 33.00 degrees
Delta/t; 5.4 degrees
Speedfan 4.33 Reading; 44.00 degrees

Sort of drives home how unreliable and inaccurate software temperature measurement is really.

Note: At load the Speedfan reading showed marked swings in temperature from 44 degrees up to 50 degrees. Water/Air ambient remained stable during these periods.

The Asus motherboard used in this HTPC build does not lend itself to overclocking mayhem so we had to settle for our default values as we tested out processor performance.

Processing performance

Intel's E8400 at default provides excellent CPU performance. It was more than capable of coping with our video encoding testing at a very respectable performance level. The E8400 was beaten out by the Q6600 in dual pass encoding trials but was more than competitive. Multi tasking using Folding @ Home whilst encoding had little effect on web browsing or other activities. We will not dwell on outright performance values of the E8400 CPU here as they are better covered by the many specialized reviews on the web.

Gaming Performance and Sound systems

We loaded a few games from the games box to test out performance. Some of the games are listed below;

COD 4
Doom 3
Bioshock
Half Life Two Ep 2
Prey
Crysis

I better start out by saying that watching Windows load for the first time on a 42" screen is something of a shock. Gaming simply compounds the visual overload. The depth and field of vision coupled with the much finer detail really add to the gaming experience.

The HTPC worked flawlessly with all of the games even with the E8400 CPU at stock settings with the exception of Crysis. Crysis presented no major problems at default settings using the onboard graphics chip. At any higher settings than default the chip complained and we were obliged to move on to a second generation card. The only other issue we encountered was with the first patch for Crysis which caused a few problems.

Sound Systems

There are a mind blowing number of sound systems out there for you to choose from when you build your first HTPC. Budget will determine your buy-in level so keep in mind the sound system is only one of the many things that can be improved later on. We used a Logitech sound system for the full theater effect. This was a major improvement in sound clarity when used with the games listed above. In fact the games were quite pleasurable to play through again for the extra benefit of the improved sound and visuals. The sound of a foot being dragged across the floor by a zombie as it approaches you from behind in the darkness was a notable major improvement. If you have played Doom 3 you will probably remember this sound as you moved through the lower levels of the Mars base.

The end result of our games testing was that we had to remove under protest the resident games guru who was totally impressed. The outcome now is that he wants to build a water cooled racing simulator.......... a more distant project perhaps.

Cables and video capture cards

Of course gaming is not everything so we did carry out extensive testing of the normal function of a HTPC namely watching movies. That is where we again hit the lack of knowledge wall. A good quality picture is dependent upon a good quality cable between HTPC and TV screen. If you are confused by the large number of cable types then take our lead here and go for a good quality HDMI to DMI cable for your HTPC creation. There are any number of sites on the web dealing with this so better read a few before making a cable choice.

Video capture cards also known as TV cards can also create issues for the novice HTPC builder. We used a Winfast product simply to learn from. There are more expensive and feature packed cards on the market but from a learning stand point these cards are good. The only issue that came up during our HTPC build was the ability of the card to recognize and load our local TV stations. A handy hint here for the Winfast card is to try every "Zone" setting as the information given for our zone was inaccurate. We managed to load all the local TV stations by simply experimenting with the zone settings.

Well some time has passed now and the HTPC has settled into its role of entertainment centre. Apart form the various competing demands made on the use of the HTPC it has been a very satisfying project from our point of view. There are still the inevitable upgrades to go in the future as technology improves so I guess we will be looking at another review of our HTPC project again as they occur.

 

Copyright Radiical Pty Ltd 2007